HPCwire: ETI Runtime System Achieves Significant Performance Improvements on Graph 500 for Intel Endeavor

Since 1986 - Covering the Fastest Computers
in the World and the People Who Run Them

Translation Disclaimer

Subscribe | Sign In

Visit additional Tabor Communication Publications

Leading HPC
Solution Providers

June 14, 2011

ETI Runtime System Achieves Significant Performance Improvements on Graph 500 for Intel Endeavor

NEWARK, Del., June 14 -- E.T. International, Inc. (ETI), a provider of system software solutions for hybrid and many-core hardware architectures, today announced significant speed improvements of the Graph 500 benchmarks on the Intel Endeavor supercomputer when using ETI’s SWARM (SWift Adaptive Runtime Machine) system software.

ETI’s SWARM runtime system improves the Graph 500 execution time by up to 11-fold over MPI, vaulting it into the top two on the current Graph 500 list. Specifically, at 256 nodes (Scale 33), it delivers 6.9 giga-edges per second (GE/s), and at 128 nodes (Scale 32), 4.2 GE/s.

“ETI has been working with Intel to deliver a runtime system meant for the most advanced many-core systems in the world,” said Rich Collier, COO of ETI. “Our proven experience in software system design for many-core heterogeneous architectures has allowed us to implement Graph 500 with SWARM to demonstrate the enabling power of our programming environment. We are very proud to prove the efficiency and speed of our SWARM technology with these performance numbers for Endeavor, and we look forward to continuing to achieve benchmark performances like these as we bring the system software to the rest of the world’s top supercomputers.”

The Graph 500 is designed to test computers’ ability to solve the complex, data-intensive problems common to social networks, international security, and medical records where they are required to store and transfer significant amounts of data in irregular, rapidly changing communication patterns. Rather than measuring the ability to perform a large number of basic arithmetic operations, the Graph 500 provides a performance metric for a more sophisticated capability to handle the increasingly complex problems presented by today’s applications.

As a software partner in the Intel-led team for DARPA’s Ubiquitous High Performance Computing (UHPC) program, ETI developed the SWARM runtime system technology for the UHPC initiative and improved the Graph 500 benchmark performance on the Endeavor supercomputer by using this system software instead of MPI. Intel’s UHPC team, jointly funded by Intel and DARPA, includes ETI Vice President of R&D Dr. Rishi Khan among its principal investigators.

“Intel Labs has been excited to work with ETI to prepare for high core count computers and systems,” said Wilfred Pinfold, Director of Extreme Scale Systems, Intel Corporation. “The challenge is to efficiently utilize those cores balancing both energy consumption and application performance. ETI's SWARM demonstrates that a new runtime based on codelets can effectively utilize those cores and will scale to billions of cores in a system.”

About E.T. International, Inc.

E.T. International (ETI) provides system software designed specifically for advanced many-core computing systems that span from supercomputers for highly scientific research to high performance portable devices for smartphones and games. As the leading many-core technology expert, ETI provides software solutions that increase performance, computing power, and efficiency so that users can fully realize the power of many-core architectures. Built on a decade of highly complex, massively parallel programming research for extreme-scale computing systems with the Department of Defense, ETI currently provides software, applications and custom engineering services for the energy, defense, semiconductor, digital media, and financial markets. For more information, visit www.etinternational.com.

-----

Source: E.T. International, Inc.

Share Options

Subscribe

Subscribe to HPCwire

Discussion

There are 0 discussion items posted.

Join the Discussion

Become a Registered User Today!

Registered Users Log in join the Discussion

RSS Feeds

Feeds by Topic

Feeds by Industry

Feeds by Content Type

Subscribe to All Content

All

May 22, 2012

May 21, 2012

May 18, 2012

May 17, 2012

May 16, 2012

May 15, 2012

May 14, 2012

May 11, 2012

May 10, 2012

Feature Articles

OpenACC Starts to Gather Developer Mindshare

PGI, Cray, and CAPS enterprise are moving quickly to get their new OpenACC-supported compilers into the hands of GPGPU developers. At NVIDIA's GPU Technology Conference this week, there was plenty of discussion around the new HPC accelerator framework, and all three OpenACC compiler makers, as well as NVIDIA, were talking up the technology.
Read more...

NVIDIA Launches Kepler Into HPC

NVIDIA has introduced its first Kepler-generation GPU product for high performance computing, and revealed some of the inner working of the new architecture. The announcement took place at the kickoff of the company's GPU Technology Conference taking place this week in San Jose, California.
Read more...

Intel Rolls Out New Server CPUs

Intel Corp. has launched three new families of Xeon processors, joining the Xeon E5-2600 series the chipmaker introduced in March. These latest chips span the entire market for the Xeon line, from four- and two-socket servers, down to entry-level workstations and microservers. A number of HPC server makers, including SGI, Dell, and Appro announced updated hardware based on the new silicon.
Read more...